Integral membrane proteins comprise about 25% of all genomes. Those that use the b-barrel motif comprise up to 4% of the genomes of Gram-negative bacteria, and are also present in mycobacteria, chlamydiae, mitochondria and chloroplasts. Exogenous proteins and peptides that self-assemble into b-sheets in membranes are also found in all kingdoms of life, including humans, where they are involved in pathogen virulence and host defense and in the action of many toxins and venoms. For example, the Anthrax toxin and the a-hemolysin of Staphylococcus aureus use b-barrels to permeabilize membranes. There are also many pore-forming peptide antibiotics that are predominantly b-sheet structure. In this ongoing research program we are pursuing a deeper understanding of the driving forces and structure-function relationships for the self-assembly of b-sheets in synthetic and biological membranes in order to design self-assembling b-sheet peptides that have interesting and useful structures and functions in membranes, and in order to identify b-barrel membrane proteins, predict their structure and engineer their function. The methods we are using for designing and engineering pore-forming peptides and for characterizing antimicrobial and cytotoxic activity have potentially important biotechnology applications in the fields of antibiotics, biosensors and drug delivery. These broad goals will be accomplished through the following lines of experimentation. Design new families of pore-forming peptides by screening libraries for members that self-assemble into pores in lipid bilayer membranes. Characterize the mechanism of pore formation in lipid vesicles and the structure of the peptide and cross characterize the antimicrobial activity of pore formers. Design and characterize pore-forming peptides by screening libraries for potent, broad-spectrum antimicrobial activity. Cross characterize antimicrobial peptides in vesicle-based systems to compare the determinants of structure and function in the two systems. Use combinatorial chemistry and high throughput screening to design peptides that self-assemble into membrane-spanning, protein-like b-barrel pores. Finally we will use genomics and proteomics to delineate the b-barrel proteomes of Gram-negative bacteria and validate b-barrel prediction algorithms. This information will be used to create a public database of potential outer membrane proteins [unreadable] [unreadable] [unreadable]